Low-complexity and high-accuracy DOA estimation for coprime arrays using Toeplitz matrices
Anh-Tuan Nguyen, Takashi Matsubara, Takakazu Kurokawa · 2017
Pal et al. proposed a method to extend a coprime array to a larger virtual uniform linear array (ULA), thus subspace-based direction of arrival (DOA) estimation algorithms can be used to detect more sources than the number of array elements. However, since the full information of the observation vector obtained by vectorizing the signal covariance matrix is not used, then applying the spatial smoothing technique, the performance of DOA estimation may be limited in some scenarios. By exploiting the full information of the observation vector, and using a Toeplitz matrix to form the covariance matrix of the virtual ULA, we propose a reduced-complexity DOA estimation method. Computer simulations reveal that the proposed method has better estimation accuracy and lower computational complexity than Pal et al.'s method.